Broadband MMIC amplifiers with IP1dB >23dBm?

For some of my projects it would be nice to have a broadband amplifier off the shelf with IP1dB>23dBm or more. Basically something like the HMC482ST89 but with greater IP1. Need to cover at least 50MHz-300MHz, with at least 15dB gain. Looking for other options gets me results which look good at first but don't actually have such high IP1 at <300MHz, or have really poor isolation (and are probably not unconditionally stable).

I'd like to stick to MMICs with internal output and input matching, but I'd be okay with an unmatched device with a good reference design for my frequency range. Any chance?

For CATV amplification or 50 Ohm system ?

50 ohm. I've seen that there are many CATV amplifiers from hittite which might otherwise do the trick...

50-300 MHz "Return Path Amplifiers" for CATV systems have those specifications and they can be easily be matched to 50 Ohm.They are very linear A-Class amplifiers and wideband matching is not very necessary, I think some TLs and C-L components are enough to match them.
Consider that..

Fair point for some reason I never considered that.... Some of them are even differential input/output which is nice for some of my applications. Can you recommend any manufacturers or models?

To be clear, I don't actually need great linearity. I'm just amplifying a narrowband carrier with fixed amplitude, and I want a high output level. But at the same time I need broadband return loss, hence why I'm looking for broadband MMICs.

For instance..
http://www.macom.com/products/produc...il/MAAM-011156

Look at also QORVO,ANADIGICS,HITTITE, etc.

I was also looking at MACOM devices, so I bought a few MAAM-011185 to try. Problem is the application circuit specifies a specific balun, the MABA-007748-CT1160, for the input and output, but this part is not stocked anywhere. It's a peculiar balun, and neither datasheet specifies what its properties are, so I can't find a replacement. I tried using a more typical 1:1 balun like the MABA-010245-CT1160, but it produces poor matching on both ports.

Can anyone deduce from the picture how this specific balun works? It looks like one winding is isolated, and the other two are coupled, but I can't tell which pins they're terminated to.

The application circuit is for 75 Ohm.Request 50 Ohm s-parameters from MACOM and design your own application circuit for 50 Ohm.They may have also a particular application circuit for 50 Ohm.
There are lots of equivalent 1:1 BalUn components and you will able to get a good match by adding some serial or parallel components.You don't have to stick to typical application circuit.I suggest you to measure input and output impedances of the amplifier first ( or use s-parameters if they are available) then design a matching circuit.You can also use single portion of the amplifier with expense of 3dB loss.

You can also build your own balun with a proper double aperture ferrite cores ( EPCOS,TDK) and some emailed copper wire.

Right, I also used 50-75ohm matching networks on the input and output. I was eventually able to get it working by adding external supply bias inductors to each output (I had thought the supply bias was provided internally by the Vcc pins.) I then built a version using LC lattice baluns to combine the ferrite baluns and matching networks. I still get the same IP1=27dbm and gain, which is good, but the output return loss is somewhat narrowband, though it may be good enough.


That's what I wanted to do, except the datasheets gave no info on the balun's construction. I believe it's a similar design to the one shown here:
http://www.brats-qth.org/training/advanced/feed2.htm
Except the middle winding is not coupled to the other two. The bizarre symbol in the datasheet still confuses me though (why put two dots on the same winding?).